Vision-Based Robotic Inspection for Mixed Aerospace Fasteners
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for inspecting and sorting mixed aerospace fasteners are time-consuming, prone to human error, and inefficient, particularly in accurately identifying and recertifying fasteners with geometric defects or deviations.
Innovation Solution
A high-speed system and method that uses vision-based robotic inspection stations with fixtureless rotary stages to optically measure and compare the profiles and features of unidentified parts against stored templates, generating identification signals for matched parts and directing acceptable parts to good areas while rejecting defective ones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional manual inspection methods are used, then operators can identify fasteners, but the process is time-consuming and prone to human error
Solution Approach 1:
The patent replaces manual mechanical inspection with an automated vision-based measurement system. The system uses optical sensors and image processing to automatically capture and analyze fastener geometry, eliminating human operators from the inspection process. This substitution enables high-speed automated identification while maintaining or improving measurement accuracy through consistent digital measurement protocols.
Solution Approach 2:
The system transforms physical fastener parameters into digital data for analysis. By converting geometric features into measurable parameters through optical sensing and image processing, the system enables rapid comparison against stored templates. This parameter transformation allows simultaneous evaluation of multiple fastener features, achieving both speed and accuracy in identification.
2Productivity
If automated vision-based inspection is implemented, then inspection speed increases, but system complexity increases
Solution Approach 1:
The system creates digital copies (templates) of known good fasteners and stores them in a database. These templates serve as reference models for comparison during inspection. By copying and storing geometric data rather than requiring physical reference samples, the system achieves rapid automated identification while managing complexity through software-based template management rather than extensive physical instrumentation.
Solution Approach 2:
The vision-based measurement system is designed to inspect multiple types of fasteners using the same hardware platform. The system can identify different fastener geometries, materials, and conditions through a single multi-functional inspection station, reducing the need for multiple specialized devices and thereby managing overall system complexity while maintaining high productivity.
3Loss of substance
If mixed fasteners are recertified, then waste is reduced, but identification reliability must be maintained
Solution Approach 1:
The system incorporates feedback mechanisms where measurement results are compared against stored templates of known good fasteners. The comparison process provides feedback on whether the inspected fastener matches acceptable criteria, enabling reliable identification and recertification decisions. This feedback loop ensures that only fasteners meeting specified geometric and material criteria are recertified, maintaining reliability while reducing waste through accurate sorting.
Solution Approach 2:
The system performs preliminary inspection and identification of fasteners before final recertification decisions are made. By pre-screening fasteners through automated measurement and comparison, the system prepares reliable identification data that supports accurate recertification outcomes. This preliminary action enables confident recertification of suitable fasteners while rejecting defective ones, reducing waste without compromising reliability.
Data Source
AI summary
A high-speed method and system for inspecting and sorting a stream of unidentified mixed parts are provided. The system includes a pair of inspection stations each of which includes a vision-based robotic subsystem including a robot configured to pick up a self-supporting part and place the part on a fixtureless rotary stage. A second subsystem optically measures a profile and features of the part during part rotation. A processor is operable to compare the profile and features of the part with the profile and corresponding features of stored templates to identify a matching template. A mechanism including a part sorter directs unidentified parts and parts having an unacceptable geometric dimension or defect to a reject part area and directs identified parts having acceptable geometrical dimensions and no significant defects to good part areas.